building illustrated on front coves
Atlanta University, Atlanta, Georgia
1 \MI - GAMBLE ROGER--, INC.
M O RTAR
Vely on Lime
-tested by time
NATIONAL LIME ASSOCIATION
NATIONAL LIME ASSOCIATION
TABLE OF CONTENTS
Properties of Mortars and order in which they ^rz discussed
St i ninth
( h i. n >i i \< I ; ii 1 1 . > ■ .
1 Wll - VAN All N
B\ 1I.IU-. K: FRED r. LEI
Leak-proof masonry laid up
with 1:3 lime mortar to
which not more than 4 bags
of cement were added to each
cubic yard of mortar.
DRY WALLS WITH LIME MORTAR
A WATER-TIGHT masonry wall has a negligible number of un-
filled joints or open spaces where brick and mortar apparently meet
but are not attached. The formation of such openings is influenced by
the properties of the materials used and class of workmanship em-
ployed. The presence of openings between units and mortar is the primary
cause of excessive water penetration into the wall. The excess water
may or may not enter into the interior of a building as a result of this con-
dition. If it does, the wet interior is a symptom of the condition. Re-
peated and prolonged efflorescence on the exterior of the wall is another
symptom. The condition, however, may exist without these symptoms.
This condition must be prevented if the masonry is to be durable. No
material, however weather resistant it may be, can endure for many years,
under conditions of excessive saturation. The most effective method of
preventing the condition, wet walls, is to lay up the masonry in a mortar
having properties which will permit the mason to completely fill all joints.
The mortar should also possess properties which permit formation of max-
imum extent of bond plus ability to maintain such bond.
The fact that openings exist is proved by the many instances of leaky
walls wherein both units and mortars are comparatively impervious. Mor-
tars of very low porosity are often associated with bad leaks. Water follows
the path of least resistance, through the openings between the masonry
materials, not through these materials.
Water that enters pores and voids of bricks, stones, mortars, etc.,
readily evaporates from the exterior surface. This is a normal condition
offering no cause for concern.
The alarming increase in number of leaking buildings during recent
years led to an extensive program of research at the National Bureau of
Standards, the Massachusetts Institute of Technology, the Mellon Insti-
tute of Industrial Research and other recognized institutions. The prob-
lem has also been intensively studied by leading authorities in Europe, par-
ticularly in Great Britain and Sweden. The data resulting from these
various studies have been analyzed and the following discussion prepared
in an effort to place authoritative information at the disposal of American
architects, engineers, building officials, builders and others.
Assuming proper design, adequate inspection during construction and
good workmanship, the most important factor in obtaining a water-tight
wall is the selection and use of a mortar properly adapted to the units,
taking into consideration the nature and severity of exposure.
Adaptability of Mortars
To prevent excessive water penetration into a wall, one must use an
adaptable mortar. This is a mortar that can be expected to adhere
satisfactorily to all types of building units.
Lime promotes mortar adaptability and a mortar rich in lime is suit-
able for all types of units. An adaptable mortar is first of all workable.
It has bonding power. It is characterized by low volume changes sub-
sequent to hardening. It has good extensibility, produces good masonry
strength and has a rate of hardening compatible with modern methods of
construction. These properties will be discussed in the light of the results
of research in the masonry field.
PROPERTIES OF A MORTAR THAT MAKE IT
ADAPTABLE AND SUITABLE
Until recent years the term, workability, has been poorly understood
in its application to masonry mortars. Warren E. Emley in Bureau of
Standards Technologic Paper No. 169 states:
"The ability to retain water, while the most important, is not the
only factor governing plasticity." Due to the fact that water retaining
capacity is the most important factor governing plasticity and further to
the fact that this property has been found to be always associated with
plasticity, it may be said that a measure of the water retaining capacity
of a mortar provides a good index to its plasticity or workability.
A factor of lesser importance is the bulk density or weight per unit
volume of the freshly mixed mortar. Still another factor is its "trowel-
ability" which may be estimated by the difficulty or ease of forcing it into
the interstices of a brick surface of very rough texture. The three known
factors that control workability are then:
1. Water retaining capacity
2. Bulk density
3. Good troweling properties
(A) Water Retaining Capacity
The method of measuring the water retaining capacity of different mor-
tars and the enhancement of this property by the substitution of more and
more lime for portland cement in lime-cement mortar mixtures are dis-
cussed in a paper, "Rate of Stiffening of Mortars on a Porous Base,"
appearing in the September 10, 1932, issue of Rock Products. These data
were obtained during the course of an extensive cooperative investigation of
masonry mortars at the National Bureau of Standards. The index to water
retaining capacity was taken as the flow of a mortar on a 10-in. flow table
after it had been subjected to suction on a standard porous base for a
definite time interval, 1 minute, 2 minutes, etc. The suction was equiva-
lent to that of a very porous dry -press brick laid dry in the wall. Water
is the sole lubricating medium in mortar and the more retentive it is of
water, the more workable is the mortar under practical working condi-
tions. If the mortar loses water to the brick too rapidly, it stiffens so
quickly that intimate contact is not made and as a consequence the mor-
tar adheres to the brick only in spots. In other words the extent of bond
is poor. If the mortar has high water retaining capacity the extent of
bond is good.
Water is transmitted readily through the unbonded areas, i.e., where
brick and mortar apparen tly meet but are not attached. This was proved
by actual tests at the National Bureau of Standards. Typical data are
given in Table 1.
Water Retaining Capacity of Mortars as Related to the
Water-tightness of Test Walls
Lime: Portland Cement: Sand
Proportions by Volume
Water Retaining Capacity
(Flow per cent, after suc-
tion for 1 minute on a
standard porous base)
Maximum rates of leakage
of 8-inch test walls made
with Dry-Press bricks
set dry. (Cubic centi-
meters of water trans-
mitted through walls per
Figure 1 illustrates what is meant by "mortar pancakes," These fre-
quently occur with mortars of low water retaining capacity. As is appar-
ent, the adhesion of the mortar to the bricks was initially very poor. The
bricks held together for at least 4 weeks but later came apart due to vol-
ume changes in the mortar occurring subsequent to its hardening.
A 1:3 Portland cement and sand mortar and porous dry-press bricks set dry. The mortar
pancake was very dense and had high strength. A wall wherein this condition exists would
have neither strength nor water-tightness.
Bricks identical to those in Figure 1 were wetted by complete immer-
sion in water for 15 minutes and then laid with this same 1:3 portland
cement mortar in making brick beams for flexure tests. The mortar hav-
ing low water retaining capacity when remaining on a base that will not
absorb an appreciable amount of water will undergo segregation and form
"puddles." The reader can see where puddles were in Figure 2 which is
typical. The bricks were found to be separated at the end of 3 months
when the beam was picked up to be tested. Volume changes in the hard-
ened mortar had again done their work. In Figure 2, practically all of
the mortar joint is seen attached to the lower brick. There is a very thin
film (cement particles mostly) over the flat surface of the upper brick.
Unbonded areas are also seen. The bond was weaker than the mortar, a
condition characteristic of mortar having poor bonding power.
It is difficult to get a satisfactory bond with a mortar deficient in water
Poor extent of hand. Porous bricks set wet I r$ minutes total immersion in water I u ith a 1:3
Portland cement and sand mortar. Note depressions in surface of mortar joint (right due
to puddles being formed by segregation in the mortar before top brick was placed. The bond
tailed completely weeks after mortar hardened. Failure was caused by volume change^
occurred subsequent to hardening of mortar.
retaining capacity, no matter what the absorption of the brick may be.
The results obtained with such a mortar and impervious bricks are com-
parable to those illustrated in Figure 2 and obtained with porous bricks
that have been thoroughly wetted before laying them. Figure 1 shows
what a porous brick laid dry with a mortar of low water retaining capacity
will do. With a brick of medium rate of absorption there is still trouble
with this kind of mortar. The mortar will tend both to segregate and
stiffen too rapidly, the more of one than of the other, dependent on the
absorption rate of the unit and the speed of bricklaying, this type of
mortar being very "sensitive" to slight variations in the absorption rate
among bricks of a kind.
The data in the last column of Table 1 were obtained by tests that were
extremely severe, far more so than the conditions to which a wall may be
exposed in the most severe rain storm. The extent of bond obtained with
the 2 lime : 1 portland cement : 9 sand and the 3 lime : 1 portland cement :
12 sand mortars in the walls subjected to these severe tests, was complete
and for all practical considerations the test walls would have been water-
tight had they been integral parts of an 8-inch wall. Practically all of the
water that came through the test walls with these mortars, followed
through along the porous header bricks. Furring would check this in a
(B) Bulk Density
The bulk density of a mortar is a factor of such minor importance that
its discussion is not warranted. Since three-quarters of the total volume
of dry ingredients is sand, there are very strict limitations imposed on any
effort to make freshly mixed mortar light in weight.
(C) Troweling Properties
A mortar has good or poor troweling properties. The data of the last
column of Table 2 were obtained with very rough textured face bricks set
wet. It was observed on breaking the test walls after testing them for
permeability that the non -plastic mortars did not fill the interstices of the
rough faces and header ends of these bricks. Rather, they tended to
bridge across these shallow indentations. The course of the water in its
movement into and through the test wallettes could be traced by the
stain left by the dye that was in the water and it was noted that water
moved under the "mortar bridges" and over the bottom of the depressions
when the bricks were laid in non-plastic mortars. The more workable
mortars gave very little evidence of this condition.
Average Rates of Water Transmission Obtained with
Different Mortars and a Rough Surfaced Brick Set Wet
Average maximum rates
of water transmission
through test wallettes.
Lime: Cement: Sand
Averages for 3 specimens.
Cubic centimeters per
These data were obtained at the National Bureau of Standards during
the cooperative study of mortars for unit masonry.
2. BONDING POWER
This property is not so intangible as its name implies. A mortar having
this property really takes hold on a solid surface. If a mortar can't take
hold, can't attach itself to a building unit under certain conditions, then
it has no bonding power under those conditions. We may strive to change
the conditions by wetting the brick, etc. However, it is a nuisance to have
to do this and moreover we never know when we have done it satisfac-
Strength is only one of the factors determining bonding power of a
mortar. Often (see Figure 1) it is a factor of small importance. We are
not concerned so much with what a mortar can be made to do with
extreme care and careful nursing in a laboratory, as we are with what it
does do with reasonable care either in a laboratory or in a wall of
masonry. With extreme care a poorly adaptable mortar of low water
retaining capacity and high volume changes subsequent to hardening may
bond well in the sense that it adheres completely to the unit at all points
of contact. In such a case the section of failure in tension is practically
always entirely at the plane of contact of brick and mortar and not in the
mortar itself. In other words, under the most favorable conditions, a
poorly adaptable mortar will always have a bond strength that is less
than the tensile strength of the mortar.
Bonding power involves general adaptability. Bonding power is
that property of a mortar which gives it a tendency to adhere uni-
formly and completely at all points of contact where bricks and
mortar meet, under widely different conditions and with various
types of units, the intensity of its adhesion being such that the
tensile strength of bond is either equal to or greater than that of
With reference to two mortars, one being a 1 lime: 1 portland cement:
6 sand and the other a 1 portland cement : 3 sand mortar (proportions by
volume), conclusion No. 6 of Bureau of Standards Research Paper No.
290, "Durability and Strength of Bond Between Mortar and Brick,"
reads as follows: "The ratio, strength of bond to tensile strength of mor-
tar, was greater with the 1:1:6 than with the 1:3 mortar." More
recent data, published in Bureau of Standards Research Paper No. 683,
"The Properties of Bricks and Mortars and Their Relation to Bond,"
show conclusively that mortars richer in lime than the 1:1:6 mix have
at 3 mos.
OtJ-tJ-O t-. CN © l£> t^OOO
o t^. o\ m
Oi oo fO m cm t-* o
- r- —
i-t es ro o
CS CM ^ ^H CM CM
q o -; o
dddd dddd o S <6 S
a o S <6
*> S; m
c » C C
.2 •£ £ £ «
(NO^-N Q0O> ©CI
O ro m t-»
O ^ m CM
m ao ffv t~» fimrsi/i t»* uo a\ N. o t-» r- cm
q o q o i-« o <-
O^ . «C u
'5 c ° 2 S
r^ m o cm
t^. CO o CN
2- - t -
^H -H f*
^H ,-1 f»
P* f* ,H
<-l ^ t-*
-^ ■« t^ J3 y
u 5 u & &? bb
1 t> '
u u £* ^ £* ^
^ « b U
T) S*OT3 *D S-OT3
T3 S -a -a
. l .
J 1/ Lj li
h — " w «:
-H « fO <£>
6 6 6c
2 2 2 2
Ratio of St
s at 3 Mon
- (S CS N CM O* Ol ON O
i£) kO lO lO
&-> ti- *r ^j- *r
TTTj- Tf Xj-
•* i- <r rr
lO o vo ^o
• i/i in i/i in r-^r^t^r-.
.E^nnn ti- tj- ^- ^r
oo m m in m
00 00 oo oc
O 00 00 00 00
iO iO ^O *£
CM CM CM CM
«m ,lM -*— •<*■
4-> -4J • 4J • «J -
c . d . c c ■
l* _,_, O K _ u
m yi/i ui/i om u
«0m0o,0 w o
o.S a£ a £ a£
w >, " * - >, " >
^^ ^ ^3 " ( * 1 JD ^ J3 '" J3
CM CM CM CN
V V i> V
O V O 4,
_. "H ""S >>- >IT1 >
-a a a a
CO CO fO CO
,11 ,U ,U ,11
_ _ _ _
« ^ p« «
better bonding power than that mortar. As lime, either putty or dry
hydrate, was substituted more and more for portland cement, the bond-
ing power was improved. The tensile strengths of mortar specimens were
not actually measured in these tests. However, the tensile strength of mor-
tar is approximately proportional to the corresponding compressive and
transverse strengths which were measured.
Table 3 contains data typical of those obtained by these tests at the
Bureau of Standards. The lime-cement mortar mixtures of Table 3 were
made with one and the same portland cement. The bonding properties
of the portland cement mortar and of the portland cement mortar with
the addition of 15 per cent by volume of hydrated lime are given as data
for the first and last mortars respectively of Table 3. The bricks listed in
this table are representative of both extreme and intermediate types from
the standpoint of rate of absorption.
The higher the values in the last 2 columns of Table 3, the
greater the bonding power of the mortar. The greater the diver-
gence among the ratios in these two columns (Table 3) obtained
with the 4 different bricks with one and the same mortar mixture,
the poorer the adaptability of that mortar and conversely.
In an article published in Part II, Vol. 33, 1933 issue of the Proceedings
of the American Society for Testing Materials, Professor W. C. Voss of
the Massachusetts Institute of Technology makes the following state-
"The position has been taken that a brick and mortar assemblage
which will fail 50 per cent or more in the mortar is far superior to one
which will not, regardless of the actual strength in pounds per
square inch. Our reason for this is based upon the fact that it is
more important to insure a 'bond layer* at the brick line, than it is to
insure no break in the mortar."
Even this concise statement gives to high strength mortars of poor
bonding power more than is their due. Note the expression, regardless
of the actual strength in pounds per square inch and then note
that the strength of bond in tension of the last mortar of Table 3 was
0.0001 times 2,185, or 0.2 lbs. per square inch with brick No. 1 (dry-
press) set dry. Low bond strength obtained with such a strong mortar is
well illustrated by Figure 1.
Data similar to those in Table 3, may be derived from the data of Duff
A. Abrams as given in Table 10 of Bulletin No. 8, Structural Materials
Research Laboratory, Lewis Institute, Chicago. Mr. Abrams tested the
bond between concrete and steel by pull-out tests of 1-inch plain steel
bars imbedded 8 inches in an 8-inch concrete cylinder. In Table 10 of
his publication he presents the compressive strength of 6 by 12 -inch
cylinders together with the maximum bond stress. In the group of tests
given in this Table (10) hydrated lime replaced an equal volume of port-
land cement ; therefore the quantity of cement decreased with increased
percentage of lime. The values in Table 4 were computed from his data.
Effect of Hydrated Lime on the Ratio, Maximum
Bond Stress to Compressive Strength
Per Cent of Cement
Plus Hydrated Lime
Maximum Bond Stress (lbs. /in.-)
Ratio: . . 9V
Compressive Strength (lbs. /in. 2 )
These values were obtained with a 1 :5 mix by volume. Lean mixtures
tend to have low bonding power and it is especially interesting to note a
tendency for slight improvement in this property with so lean a mix with
the substitution of lime for cement.
With reference to reinforced brick masonry, Professor John W. Whitte-
more makes the following statement in Vol. XXV, Bulletin No. 9, of the
Virginia Polytechnic Institute:
1 'Failure of the individual bricks in tension, bending or shear is
much less likely to occur than is the failure of the mortar joint in
tension, bending or shear. The adhesive strength of mortar to
brick is perhaps the governing factor. 11
On page 47, Vol. Ill, Brick Engineering, Hugo Filippi states:
"From the general performance of beams and slabs under test,
however, there is some evidence that the bond value of the mortar
joint in shear is an important, if not the controlling factor/ 1
We have seen that to have bonding power, a mortar must have adapt-
ability. To have adaptability it must have workability. Good bond neces-
sitates good extent of bond. Good extent of bond under variable condi-
tions and with different types of units can only be obtained with a mortar
that is adaptable.
architects: delaxo and aldrich
Divinity School— Vai.i
urn. i n "l
tight masonry—free from efflorescence
volume cement and ». Vv , ^Pose
gUflflisiTY, New Haven, Connecticut
ytstfli' Mortar composed of two volumes lime, one
r , , . nine volumes sand.
GENERAL CONTRACTORS: SPERRY AND TREAT CO.
3. VOLUME CHANGES
There has been more confusion in the treatment and consideration of
this topic than perhaps any other. It must be borne in mind that mortars
undergo three distinct types of volume changes. These are:
(A) Compacting on a porous base
(B) Shrinkage during early hardening
(C) Volume changes subsequent to hardening
The first two, (A) and (B) occur but once during the life history of a
mortar. The third, (C) occurs over and over again. It is a cyclic process,
the mortar expanding on becoming damp or wetted and shrinking when
it again dries out. This cyclic process, alternate expansion and contrac-
tion, continues over a period of years. The first type of volume change,
(A), cannot take place unless there is contact of the mortar with another
material, i.e., a porous body. The types (B) and (C) do not require con-
tact of mortar with a solid object.
(A) Compacting on a Porous Base
a. This type of volume change is essentially unidirectional. The dimi-
nution in volume is caused chiefly by a shortening of one dimension of the
mortar joint and in a direction perpendicular to the plane of contact of the
mortar and solid object, such as brick, tile, etc. The shrinkage is in the
general direction of the course that the water takes in leaving the mortar
and entering the porous unit. The higher the water retaining capacity
of the mortar the less is its compacting on a porous base and con-
b. Compacting on a porous base occurs prior to any appreciable
hardening or cementing action that the mortar undergoes.
c. It takes place in a very few minutes and is most rapid at the first
instant of contact.
d. It is attended by a rapid stiffening of the mortar, a rapid loss of
workability due to rapid loss of water. This stiffening is not harden-
ing through cementing action and must not be confused with it.
e. The rapid stiffening of a mortar due to compacting on a porous base
renders it impossible to make good contact between the mortar bed and
the next course of bricks or to slush the mortar in cross joints when shov-
ing bricks into place. It prevents the mortar in the horizontal bed on a
course of bricks from slumping into and filling any vertical joints left open
by the mason.
f. The unidirectional compacting pulls the mortar away from the bricks
in the cross joints. The mortar is too stiff to flow under the pressure of
bricks laid above it to cause any renewal of contact.
Lime and rich in lime mortars have high water retaining capacity and
therefore tend to undergo a minimum of compacting when in contact
with porous building units (see "Rate of Stiffening of Mortars on a
Porous Base," Rock Products, September 10, 1932 issue).
(B) Shrinkage During Early Hardening
a. This type of volume change is omnidirectional.
b. It attends hardening through cementing action.
c. It occurs but once during the life history of a mortar.
d. Its magnitude is diminished when the mortar is in contact with an
absorbent unit and is greatest when in contact with solids of zero porosity.
e. A mortar which hardens slowly is more or less plastic when most
of this type of shrinkage is completed.
f. Any possible damaging effect of this type of volume change is de-
pendent as much (and probably more) on the rate of hardening as on
the magnitude of shrinkage.
g. Data obtained during an extensive study at the Bureau of Standards
show conclusively that this type of volume change is not damaging either
to the mortar or to the bond between the mortar and bricks. This was
found to be true regardless of the magnitude of the shrinkage of mortar
during early hardening. Typical data are given in Table 5. Had the
shrinkage during hardening been damaging, then with mortars relatively
high in such shrinkage, we would expect that the strength of bond when
Negative Results Obtained in the Study of Any Possible Damaging
Effect of Shrinkage During Early Hardening
(Initial 48 Hours)
of Bond in
Tension at 3 months
of 3 Tests
Lime: Cement: Sand
Brick No. 3
Brick No. 5
lbs. /in. L>
lbs. /in. 2
lbs. /in. -
lbs. /in. 1
Photo by Arthur C. Haskell
Detail of brickwork showing condition of brick and mortar after more than 2<o years of
exposure %n severe New England climate.
Joseph Peaslee Garrison House, Rock Village, Massachusetts
BUILT 1675 — PHOTOGRAPHED 1 933
Lime mortar used,
metal lugs were embedded in the joints would be very much less than that
obtained without the lugs and that this difference would increase with
the magnitude of such shrinkage. This did not happen. The data were
obtained by bonding together the flat surfaces of two bricks with half-
inch mortar joints. Three %-inch lugs were embedded in the joints of
half of the total number of the two brick-mortar specimens.
In table 5, most of the values obtained with lugs are slightly lower than
those obtained without lugs but this difference was apparently independ-
ent of the magnitude of the shrinkage during early hardening. The
slightly lower values obtained with lugs are explained on the basis of the
fact that where the metal lugs touched bricks, mortar did not. There was
slightly less bonded area with the lugs present.
Bricks 3 and 5 (Table 5) were about as impervious as any that could
be found, a condition designed to augment the shrinkage during early
hardening. The bricks were restrained from coming together as the mor-
tar shrunk, a condition similar to that obtaining in the vertical joints of a
wall. An intensive study of available information does not reveal any
authentic data obtained by carefully controlled tests, that is in conflict
with the conclusion that shrinkage during early hardening does not
create openings between mortar and bricks.
(C) Volume Changes Subsequent to Hardening
a. This type of volume change is characterized by expansion on wetting
and shrinkage on drying.
b. It is cyclic and frequently is rapid if the conditions for wetting and
drying are favorable.
c. It takes place in rigid mortar that has an elastic limit.
d. High volume changes subsequent to hardening are characteristic of
dense mortars of outstanding hydraulic properties. They are lowest for
straight lime mortars. This type of volume change is reduced as lime is
substituted more and more for portland cement (Bureau of Standards
Research Paper No. 321, "Volume Changes in Brick Masonry Materials").
The data shown in Table 13 of Bureau of Standards Research Paper
No. 683, "Properties of Mortars and Bricks and Their Effect on Bond,"
show that when the extent of bond is poor (less than 90 per cent of the
flatside area of a brick) there is a tendency for it to be destroyed entirely if
the mortar is one that has high volume changes subsequent to hardening.
If the extent of bond is complete or nearly so, high volume changes
in the hardened mortar tend to produce cracks in a direction more or
less perpendicular to the plane of contact of brick and mortar. These
cracks may heal slowly when the mortar is wet, but they occur again and
again and tend to admit an excessive amount of water into the wall and,
as a consequence, further occurrence of volume changes is accelerated.
Usually, a mortar having low water retaining capacity is one that also
undergoes relatively high volume changes subsequent to hardening. This
is an undesirable combination, for it usually results in a poor extent of
bond coupled with conditions which tend to completely destroy the bond
and leave "mortar pancakes" in the wall.
Extensibility is here denned as the maximum linear deformation
Extensibilities and "Factors of Safety" of Mortars,
Age 3 months — Average of 3 tests
of second col-
Lime: Cement: Sand
by the corre-
values of the
0:1:3 (portland cement No. 1) (a)
0:1:3 (portland cement No. 2) (b) . .
0:15:1:3 (portland cement No. 1 and lime
1 :1 :6 (portland cement No. 1 and lime No. 1 )
2:1:9 (portland cement No. 1 and lime No. 1 )
3:1:12 (portland cement No. 1 and lime No. 1 )
1:0:3 (lime No. 1) (c) . .
1:0:3 (lime No. 2) (d)
1:0:3 (lime No. 3) (e)
1:0:3 (lime No. 4) (f)
1 :1 :6 (portland cement No. 2 and lime No. 1 )
2 :1 :9 (portland cement No. 2 and lime No. 1 )
3:1:12 (portland cement No. 2 and lime No. 1 )
1:1:6 (portland cement No. 1 and lime No. 3)
2:1 :9 (portland cement No. 1 and lime No. 3)
3:1:12 (portland cement No. 1 and lime No. 3)
Portland Cement No. 1— Typical Grey Portland Cement meeting A.S.T.M.
Portland Cement No. 2— White Portland Cement meeting A.S.T.M. Standard
Lime No. 1— Quicklime meeting A.S.T.M. Standard Specifications.
Lime No. 2— Hydrated Lime meeting A.S.T.M. Standard Specifications.
Lime No. 3^Hydrated Lime meeting A.S.T.M. Standard Specifications.
Lime No. 4— Quicklime meeting A.S.T.M. Standard Specifications.
(stretching) that a hardened mortar may undergo before it fails in ten-
sion. It is a dimensionless value and may be expressed as inches per inch,
inches per one hundred inches, etc. It happens that this property does
not vary a great deal among mortars of different compositions, provided
that the proportions of cementing materials to sand is 1 to 3 respectively,
by volume, in all cases. In tests made during the cooperative investiga-
tion at the Bureau of Standards, the extensibilities of 50 different mortar
compositions ranged from 0.014 to 0.035 per cent. The bulk of the speci-
mens tested had extensibilities within the limits, 0.020 to 0.030 inches per
100 inches. These values were obtained by transverse tests and the exten-
sibility was computed by dividing the modulus of rupture by the modulus
of elasticity, it being assumed that failure was in tension, the mortar speci-
men acting as a beam supported at both ends and loaded in the middle.
The real significance of the property, extensibility, is its relation to the
magnitude of volume changes subsequent to hardening. If the hardened
mortar shrinks 0. 10 of an inch per 100 inches and has an extensibility of 0.02
of an inch per 100 inches, its "factor of safety" would be 0.02-0.10 or I.
If the shrinkage of the hardened mortar is 0.01 of an inch and the exten-
sibility is 0.02 of an inch per 100 inches of mortar, the factor of safety is
.02-^.01, or 2. The latter condition is surely more desirable than the
A factor of safety of 1 or greater (Table 6) is a desirable condition.
With mortars of values appreciably less than unity there is likely to be
either cracking open of joints (if the extent of bond is good) or shearing
loose from the bricks or other units (if the extent of bond is poor). This
danger is obviated by the use of lime or lime-portland cement mixtures
that contain enough lime to keep the ratio of extensibility to maximum
shrinkage of the hardened mortar above unity. Such mortars remain
bonded to building units for years and few if any shrinkage cracks appear.
Table 6 contains data typical of those reported by the Bureau of Stan-
dards in the cooperative mortars and masonry investigation. (National
Bureau of Standards Research Paper No. 683.)
A strong wall is one having integrity, i.e., adhesion of mortar to units
throughout. "Mortar pancakes" that keep bricks apart and not together
do not provide a strong wall, be the strength of the mortar what it may.
A chain is only as strong as its weakest link. As we have seen this weakest
link is at the plane of contact of bricks and mortar when the latter has
poor bonding power. All other things being the same, a water-tight wall
is always stronger than a leaky one. Under normal conditions a water-
tight wall will satisfactorily meet all ordinary requirements in the matter
of strength in so far as masonry of the average load bearing type is con-
The following comments relative to strength of mortar are to be found
in circular No. 30 of the National Bureau of Standards:
"This question of the strength of a mortar is apt to be given undue
weight. Since masonry is assumed to weigh 150 pounds per cubic
foot, then the compressive load (in pounds per square inch) at the
bottom of a wall will be Hi times its height in feet. A mortar with
a compressive strength of 100 pounds per square inch, should, ac-
cording to this reasoning, be able to carry a wall 100 X \U = 96
feet high, or about nine stories. The compressive strength of the mor-
tar is usually measured by crushing 2-inch cubes. For a homogeneous
material the unit compressive strength varies with the shape of the
specimen, being dependent upon the ratio between the least hori-
zontal dimension and the height. In a cube, this ratio is one. A mor-
tar joint in a wall may possibly be 9 inches wide by 30 feet long by
Vz inch thick. In this joint the ratio is 9 - Vz = 18. If a mortar has
a strength of 100 pounds per square inch when tested in the form of
a cube, it should theoretically have a strength of 1800 pounds per
square inch when laid up in the wall."
Much consideration has been given to the compressive strength of brick
(Data typical of those in Table 13, National Bureau of Standards Research Paper No. 683)
Transverse Tests of Brick Beams (5 bricks, single tier, 4 intervening mortar joints)
Values are Averages of Three Tests. Beams Aged Three Months.
Average Modulus of Rupture
Lime: Cement: Sand
1:1:6 (Lime No. 1)
1:1:6 (Lime No. 2)
2:1:9 (Lime No. 2)
3:1:12 (Lime No. 1)
1:0:3 (Lime No. 2)
lbs. /in. 2
lbs. /in. 2
lbs. /in. 2
lbs. /in. -
lbs. /in. 2
lbs. /in. -
'Poor extent of bond obtained with at least one of the 3
Woman's Ward, Western State Hospital, Fort Steilacoox, Wash.
architects: heath, gove & bell, mock & morrison.
Mortar proportions were two volumes lime putty, one volume Portland cement and
nine volumes sand.
piers. It is interesting as well as enlightening to consider the results of
flexural tests. On page 67 of "Impervious Brick Masonry," publication
by the Alton Brick Co., St. Louis, Mo., the results of some flexural tests
with two brick-mortar specimens are given. The average modulus of
rupture was 41.9 lbs. per square inch with portland cement mortar and
79.7 lbs. per square inch with a 1 lime: 1 portland cement: 6 sand mortar,
using the same brick (6 per cent total absorption) in both cases.
A poor bond is not as noticeable in compression as it is in flexural tests.
There have been far less data obtained by the latter than by the former
tests. Unfortunately all too many investigators have overlooked the fact
that a brick pier of relatively enormous compressive strength may also
be one in which the extent of bond is poor throughout. The same pier
when subjected to a test for permeability will often leak like a sieve.
The data of Table 7 are taken from those of Table 13 of National
Bureau of Standards Research Paper No. 683, "The Properties of Bricks
and Mortars and Their Relation to Bond."
The reader may note that widely divergent values were obtained with
the portland cement and the portland cement plus 0.15 volumes of lime
mortars. The average modulus of rupture of the beams with the portland
cement mortar and Brick No. 3 (of very low absorption) was the same as
that obtained with the lime mortar and this brick. This was due to poor
extent of bond in the former case.
Brick No. 5 (practically zero absorption and of a glassy surface) gave
low bond strength with all mortars. However, with the more workable
mortars, the extent of bond was good with this brick.
Some of the individual values obtained with the first two mortars
(Table 7) were zero, owing to the fact that when the time came for testing
the beams it was found that these mortars, poorly bonded initially (due to
segregation in the mortar), had come loose from the bricks during the
aging period. The same beams were intact during the first 4 weeks, hence
it is concluded that the bond in these cases was disrupted through the
relatively high volume changes subsequent to hardening, characteristic
of these two mortars. The poor extent of bond in Figure 2 is an illustra-
tion of the functioning of the first mortar of Table 7 in the beam tests.
Many authorities recommend that porous bricks be wetted to improve
the extent of bond. In obtaining data of Table 7, the porous bricks, Nos.
1, 2, 4 and 6 (ranging from medium to high absorption) were all wetted
by total immersion for 15 minutes just before laying them. The mortars
of low water retaining capacity tended to se^re^are even though they
could not lose water through brick suction. High water retaining capacity
is always essential with all types of units. Moreover, mortars deficient
in this property are usually characterized by relatively high volume
changes subsequent to hardening. These two undesirable properties, low
water retaining capacity and high volume changes subsequent to harden-
ing, arc usually associated. The same holds for the desirable combina-
tion, high water retaining capacity and low volume changes subsequent
to hardening, and this desirable combination does more to promote
flTJ ,Z* WaH ° f maS ° nry than moTtar stre ^ th P«" ~. This
InTjt 1 apPre ° ated and thoroughly borne in mind by all who are
s^ssssr* unit masonry from the standp ° int ° f b ° th strength
The greater the divergence of the average values of Table 7, with any
bt'v of thTr" f " tH ; 6 ^^ ° f brick ' the P°°- «• the adapt"
zero to 208 m -nd conversely. For example, values range from
and rom 1 " tT^lT "* "* ** St ™* ht P ° rtland cement mortar
and from 15.2 to 42.1 lbs. per square inch with the 3 lime: 1 portland
cement: 12 sand mix. This checks the data of Table 3 and with a different
type of test specimen.
Bricks have received more than their just share of blame for the trouble,
lack of bond. The fact that today there are more extreme types of build-
ing units on the market than obtained in times past, renders it all the
more necessary that only adaptable mortars be used. As we have seen
from actual data, an adaptable mortar is an important factor in obtaining
strength of masonry.
School Board Administration Toweh and Centkal School Building
architects: heath, gove & bell contractors: f. h. goss construction ( omfani
Mortar proportions were two volumes lime putty, one volume cement, and nine volumes sand.
Mortar in excellent condition after more than jo years of weathering.
6. WEATHER RESISTANCE
Two factors control the resistance to frost in unit masonry. These are:
1. The extent to which the wall becomes saturated during its life
2. The frost resistance of the building materials (units and mortar).
The first of these two factors is much more important than the second.
A comparatively dry wall is not exposed to weathering from severe cli-
matic action. Water is the universal solvent and chief weathering agent.
When water which saturates a wall is repeatedly frozen and thawed the
masonry is disrupted regardless of the fact that the materials composing
it may, in themselves, be frost resistant in the usual laboratory tests. The
extent of bond between units and mortar in such a wall is poor to begin
with, else the masonry would not be excessively saturated. A poorly dis-
tributed bond (not complete in extent) soon fails entirely from frost
The freezing and thawing of moisture which, under the most severe con-
ditions, only dampens or partially saturates the wall, does relatively little
damage. In such a wall, the extent of adhesion between mortar and units is
good. The bond is therefore relatively durable both because it is uniform
and complete and because it is subjected to a minimum of exposure.
It so happens that those mortars which have the best freezing and
thawing records in laboratory tests are usually associated with leaky
masonry. The properties that enable them to endure severe freezing and
thawing in the laboratory are not properties which cause them to bond
completely and firmly to various types of building units. It also happens
that those mortars of relatively less resistance to laboratory freezing and
thawing are usually associated with dry walls. They have properties
which insure good adhesion at all points throughout the wall. It is more
important to avoid excessive exposure of a wall than it is to have ma-
terials that are the most resistant when severely exposed. Immunity from
the condition, excessive wall saturation, is more essential than the ability
of the materials composing the wall to endure excessive satura 1
ssive dampness in itself is damaging both to health and property.
Lime mortar has an enviable record for resistance to weathering in the
oldest brick buildings in the history of this country. The walls in these fine
old buildings seldom if ever became completely saturated. Therefor'
lime mortar did not decay and fall out.
Due regard and care should be given to the subjects, design, construe
tion, workmanship and maintenance. Masonry should not be too freely
Photo by Arthur C. Haskell Courtesy of Pencil Points
Hollis Hall— Harvard University, Cambridge, Massachusetts
built i763 — photographed 1 933
Lime mortar with a record of 170 years of service in a severe climate.
exposed, walls should be furred, parapet walls should be protected and
sills, copings, etc., should be properly flashed. However, attention to
these details alone will not prevent excessive saturation of a wall.
With these precuations there must be a judicious selection of mortar ma-
terials on the basis of the results of exhaustive research in the field of mor-
Lime and rich in lime mortars attain their full strength at a rate less
rapid than that of portland cement mortars. Many of the natural cements
and hydraulic limes also attain their full strength at rates comparable to
that of lime mortars. To subject specimens of mortars having a relatively
slow rate of attaining strength to severe laboratory freezing and thawing
tests at an early age is a procedure that can only add to an accumulation
of "misinformation" which is of no practical value. The number of cycles
of freezing and thawing provided for in the usual laboratory procedure
before specimens have aged a year is more than would be attained, even
in a leaky wall, in half a century of normal exposure. The proof of this
statement is found in the many examples of old masonry in both the
United States and Europe made of materials which of themselves give
negative results under prescribed laboratory tests, yet these buildings
stand today in mute testimony of their endurance after centuries of ex-
All building materials contain at least some trace of water soluble sub-
stances that can appear on the exterior wall surface if there is excessive
saturation of the wall.
Table 1 of National Bureau of Standards Technologic Paper No. 370,
"Cause and Prevention of Kiln and Dry -House Scum and of Efflorescence
on Face-Brick Walls," shows the superiority of lime mortar from the
standpoint of water soluble salt content.
Use lime and keep the wall dry.
J. A. Van der Kloes, in an article, entitled "Influence of the Qaulity of
Mortar on Masonry," Quarry, 1911, 16, pages 204-7, states that he does
not "remember ever having seen efflorescence on brickwork set in lime
mortar of only lime, fat or hydraulic, and sand."
M. Hasak, in "Was der Baumeister vom Mortel Wissen muss," pages
67-69, Berlin, 1925, Kakverlag, G. m.b.H., deprecates the use of cement
mortar on account of its liability to cause efflorescence.
In the British Building Research Station Special Report No. 18, "The
RATE OF HARDENING
Weathering of Natural Building Stones/' by R. J. Schaffer, there is found
on page 63: "In the experience of the Building Research Station, cement
mortars are found to be somewhat more liable to cause efflorescence than
lime mortars, although hydraulic lime mortars are not always entirely
These statements are not opinions. They are founded on facts. More-
over, the use of rich in lime mortars promotes dry walls and in so doing
tends to avoid efflorescence.
8. RATE OF HARDENING
Some have expressed skepticism about the use of lime or rich in lime
mortars for the reason that they believe that such mortars harden too
slowly to meet the needs of modern construction. Let the facts dispel
such an illusion.
Rates of Hardening of Mortars on an Impervious Base
as Indicated by Vicat Tests
Rates of hardening. Averages of
3 tests. Values obtained with a
modified vicat apparatus after
Lime: Portland Cement: Sand
the mortar had been on an im-
(Proportions by Volume)
pervious (steel) base for 2 hours.
Millimeters penetration in 30
0:1:3 (portland cement No. 1)
0:1:3 (portland cement No. 2)
1:1:6 (portland cement No. 1, lime No. 1)
2:1:9 (portland cement No. 1, lime No. 1)
3:1:12 (portland cement No. 1, lime No. 1)
1 :1 :6 (portland cement No. 2, lime No. 2)
2:1:9 (portland cement No. 2, lime No. 2)
3:1:12 (portland cement No. 2, lime No. 2)
0.15:1:3 (portland cement No. 2, lime No. 2)
1:1:6 (portland cement No. 1, lime No. 4)
2:1:9 (portland cement No. 1, lime No. 4)
3:1:12 (portland cement No. 1, lime No. 4)
0.15:1:3 (portland cement No. 1, lime No. 4)
1:0:3 (lime No. 1)
1:0:3 (lime No. 2)
Portland Cement No. 1— Typical Gray Portland Cement meeting A.S.T.M. Stand-
Portland Cement No. 2— White Portland Cement meeting A.S.T.M. Standard
Lime No. 1 — Quicklime meeting A.S.T.M. Standard Specifications.
Lime No. 2 — Hydrated Lime meeting A.S.T.M. Standard Specifications.
Lime No. 4— Quicklime meeting A.S.T.M. Standard Specifications.
v- co C
o c c
u . .
o p c
fO PO CO
6 o ©
° -c ~
•n u C
S 3 °
73 co -r,
O t- p
*J: <^ ^
^ O «
u u 3
aj >. D*
"2 8 S
o ffi w
u e t -
M « P
- U <L>
"O c ^
k o to
.S > 5
u 4J CO
O ffi W
(N <N M
u C t *
^ U D
*o c J*
tx O CO
u ° 3
co >> cr
O ffi W
W) T3 cO
c c a
c3 i^ r-
QJ k 5
u c s
.2 co co
x cc g
^ * E
Lime mortar containing no cement does harden slowly on an impervious
building unit. With more porous units such as dry -press bricks there
should be no difficulty whatsoever. Lime mortar is particularly suitable
with very porous units set dry in winter construction. The bulk of the
water is out of the mortar and in the brick before it can be frozen. At the
same time the lime mortar is sufficiently retentive of water to make the
necessary intimate contact with the highly porous units without necessi-
tating wetting such units to any extent whatsoever. Bricks should never
be wetted when laid in freezing weather.
Table 8 gives typical data obtained during a cooperative investigation
as recorded in National Bureau of Standards Research Paper No. 683.
Complete data covering 50 mortar compositions studied may be ob-
tained from the National Lime Association, Washington, D.C., on request.
The higher the value of the second column, Table 8, the more slowly did
the mortar harden on the impervious base.
From the foregoing study of essential mortar properties it is apparent
that lime mortar meets all of the requirements, including strength of
masonry, for most purposes.
However, in order to avoid delay, where rapid methods of modern con-
struction are employed, the addition of portland cement to lime mortar is
recommended, even though this practice adds to the cost of construction
and often results in considerable sacrifice in quality of the finished
masonry. When cement is so added it replaces an equivalent volume of
lime in the mortar mixture.
It must be remembered that as portland cement is substituted
more and more for lime, there is a greater and greater sacrifice of
the essential properties, water retaining capacity, workability,
adaptability, bonding power, and low volume changes subsequent
Laboratory and service tests have both shown that a mortar consisting
of two volumes of lime, one volume of portland cement and nine volumes
of sand is well adapted for use with a wide variety of units and under a
wide variety of conditions, it is therefore recommended for general use
where a mortar having a rapid rate of hardening is required. If it is de-
sired that the mortar strength be increased, it is recommended that the
sand content be reduced to 7 or 8 volumes but that the ratio, 2 volumes
of lime to 1 volume of portland cement, be maintained. This procedure
will increase mortar strength with no sacrifice of the more essential prop-
For walls and piers below grade and for masonry that is continuously
exposed to wet or damp conditions, a mortar consisting of one volume
each of lime and portland cement together with six volumes of sand is
Table 9 is a list of recommended mortar mixes for various types of
masonry, in different locations and under different loading conditions.
These data should be useful as a reference in the drafting or revision of
municipal, state and sectional building codes, and in the preparation of
specifications for engineering and building construction.
CONSTRUCTED IN I 847
Common brick bonded
with straight lime mortar.
A fine example of the en-
during quality of masonry-
laid up in lime mortar.
THOMSEN I ELUS CO